STUDY EXPOSED TO CADMIUM OXIDE FUME - oem.bmj.com · of each specimen of urine received was estimated by polarography. The demonstration of protein in the urine is an objective test

Brit. J. industr. Med., 1959, 16, 135.

A FOLLOW-UP STUDY OF MEN EXPOSED TOCADMIUM OXIDE FUME

BY

J. A. BONNELL, G. KAZANTZIS, and E. KING

From the Department for Research in Industrial Medicine (Medical Research Council), The London Hospital

(RECEIVED FOR PUBLICATION JUNE 30, 1958)

The results of a follow-up study of 100 men first examined in 1953 after exposure to cadmiumoxide fume are presented and discussed.

In 1953 there were 19 cases of chronic cadmium poisoning amongst these men. Twenty-fournew cases are now described. The first signs of chronic cadmium poisoning could be seen after alatent interval after exposure had ended. Respiratory function tests showed a greater deteriora-tion in performance with increase in age in the exposed groups compared with the control group.The results in individual cases of clinical and radiological examination and of the respiratoryfunction tests showed a deterioration in the condition of those men with emphysema in theoriginal survey. and took place despite the fact that the majority of the men with chronic cadmiumpoisoning were not further exposed to cadmium after 1953. Eighteen of the 24 new cases hadproteinuria only.The evidence that chronic cadmium poisoning is associated with renal damage is discussed.

Chronic cadmium poisoning is an insidious diseaseassociated with the development of emphysema andthe excretion in the urine of a characteristic proteinof low molecular weight. Emphysema and pro-teinuria may both be present or may occur separatelyin a particular case (Friberg, 1950; Bonnell,1955).

In 1953 the health of 100 men exposed to cadmiumfume during the manufacture of copper-cadmiumalloys at two factories in England was investigated(Bonnell, 1955). As a result of that survey 19 casesof chronic cadmium poisoning were describedamongst the men then at work, and four men whowere so disabled by their symptoms as to be unableto work were examined in hospital, making a totalof 23 cases. Of these 23 men, 11 had emphysemaand proteinuria, eight had proteinuria alone, andfour had emphysema alone. One man died inuraemia, the result of chronic renal failure.

Reports of further cases of proteinuria at one ofthese factories led us to undertake a follow-up studyof the exposed men in 1957, four years after theoriginal survey. It was hoped to obtain informationabout the course of the disease, particularly onwhether it was self limiting in the absence of furtherexposure to cadmium, and whether the firstsymptoms or signs of poisoning could occur aftera latent interval.

Method of InvestigationIn order to preserve continuity with the initial investi-

gation in 1953 the two factories will be referred to asFactory A and Factory B; the exposed groups at thesefactories, Group IA and Group lB respectively; andindividual patients will again be referred to as Case A(n)and Case B(n). All the men comprising Groups IA andIB who could be traced were asked to take part in thesurvey. Each man was interviewed and a clinical examina-tion made. As in the previous survey the occupationalhistory, past medical history, and family history wereobtained, and specific enquiry was made about waterydischarge from the nose, anosmia, yellow discolorationof the teeth, cough, dyspnoea, and frequency of micturi-tion. At the clinical examination, blood pressure andchest expansion were measured. The haemoglobin con-centration, erythrocyte sedimentation rate, haematocrit,and plasma proteins showed no abnormality in 1953 andthese estimations were not repeated.

All the men provided early morning specimens ofurine and additional specimens were obtained when themen attended for examination: 24-hour specimens wereobtained from as many of the workmen as possible.The early morning specimens were tested for specificgravity, sugar, and blood, and all specimens were testedfor protein. The methods used were (a) the boiling test,(b) 250% trichloracetic acid, (c) 30% sulphosalicylic acid,and (d) "altest" reagent tablets (supplied by Ames Co.Ltd.-active principle sulphosalicylic acid). Esbach'squantitative test was carried out on all urines givingpositive results with any one test for protein. Two

135

on 2 August 2019 by guest. P

rotected by copyright.http://oem

.bmj.com

/B

r J Ind Med: first published as 10.1136/oem

.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/

BRITISH JOURNAL OF INDUSTRIAL MEDICINE

methods were used to test for glucose, one of which wasa specific qualitative test and the other quantitative.Dipsticks ("clinistix", Ames Co. Ltd.), utilizing anenzymatic reaction, were used on all specimens of urine;a positive reaction is obtained at very high dilutions ofglucose and the reaction is specific. All specimens inwhich a positive reaction was obtained by this methodwere then tested quantitatively by the "reagent tablet"("clinitest", Ames Co. Ltd.). The cadmium concentrationof each specimen of urine received was estimated bypolarography.The demonstration of protein in the urine is an

objective test which can be confirmed by testing repeatedspecimens of urine. The presence of the characteristicprotein in the urine is taken as sufficient evidence for thediagnosis of chronic cadmium poisoning.

Respiratory function tests designed to show impair-ment of the ventilatory function of the lungs were per-formed by all the men who attended for examination.The following measurements were made: (a) Curves ofexpiratory fast vital capacity (forced expiratory spirogram,Gandevia and Hugh-Jones, 1957). Three curves wereobtained from each subject and the mean time constantdetermined. (b) Maximum ventilatory capacity at 70respirations per minute (M.V.C.70) (maximum voluntaryventilation 70, Gandevia and Hugh-Jones, 1957). Twoestimations were made, and the mean value determined.(c) The swept fraction was determined by expressing themean tidal excursion in the maximum ventilatory capacitytest as a percentage of the vital capacity. (d) The vitalcapacity was taken as the mean of the three expiratoryfast vital capacity volumes.The tests were again performed on the spirometer

described by Bernstein, D'Silva, and Mendel (1952),and by the same operator as before. The details of theconduct of the tests and of the measurement of thetracings have already been described (D'Silva, Freeland,and Kazantzis, 1953; D'Silva and Kazantzis, 1954;Kazantzis, 1956).A new control group was made up of the total comple-

ment of men working in the brass foundry in Factory B.There were 63 men in this group. They gave occupationaland past medical histories, and all performed therespiratory function tests. The original control groupscould no longer be examined, as many of the men hadfound alternative employment since 1953. All the menseen on this occasion were asked to attend for chestradiographs. As on the previous visit to these factoriestwo postero-anterior chest views were taken of eachman, one in full inspiration and one in full expiration.The chest radiographs were read by two observers who

were not aware of the identity of the patient when theradiographs were viewed. The readings were repeatedthree times using the same technique as before. Thirty-five of the 46 men examined at Factory A and all 37 ofthe men examined at Factory B attended for chestradiological examination. The radiological findingswhere relevant are included in the case notes (Appendix 1).As in the previous investigation independent assess-

ments of the respiratory system were made by clinicalexamination, radiological examination, and by tests ofrespiratory function. The diagnosis of emphysema was

made by taking into consideration the clinical findings,.the radiological findings, and the results of the respiratoryfunction tests. Unless two of these three criteria weresatisfied no man was regarded as suffering fromemphysema.The cadmium concentration of the atmosphere in the

casting shops at both factories was estimated continuouslyover a period of five working days. Two methods wereused. (a) For quantitative atmosphere sampling abenzamide pad (60 mesh), I inch in diameter and10 mm. in thickness was used as a collecting medium.An Edward's IV pump was used to draw the air throughthis pad at an air flow rate of 10 litres per minute. It hasbeen our experience that the resistance of the benzamidepad at this rate of air flow is constant, enabling a flow-controlling orifice to be inserted between the pad andthe pump. Before reaching the benzamide pad the airpassed through an elutriator of the type described byWright (1954), but designed in this case to remove all thepartizles of cadmium oxide greater than 7 ,u in diameter.(b) In addition, a strip recorder was used to establish thetimes of greatest emission of cadmium oxide fume.

ResultsIn the two factories 83 of the 100 men who com-

prised Groups IA and lB were re-examined. In all,24 new cases of chronic cadmium poisoning werediscovered, five with emphysema and proteinuria,one with emphysema alone, and 18 with proteinuriaalone. Thus of the 100 men first seen in 1953, atotal of 43 are now known to have evidence ofchronic cadmium poisoning. There were in additionfour men whose case records were described byBonnell (1955), namely, Cases Al, A2, A3, and A4,and two men formerly employed in Factory A andseen for the first time in consultation with their-family doctors in May, 1957 (Cases Dl and D2,Appendix 1).

Factory A.-Forty-six of the 58 men who com-prised Group IA attended for examination. Of the12 absentees, four were known cases of chroniccadmium poisoning, Cases A6, AlO, and A13, whowere unable to travel to the factory because of severedyspnoea, and Case A9 who had died of carcinomaof the bronchus. One man refused to attend for afurther examination and seven had left the company'semploy. Five of these were reported by their generalpractitioners to be in good health and two could notbe traced.Of the 46 men seen on this occasion in Group IA,

there were 19 cases of chronic cadmium poisoning;14 were new cases (Table 1) of which five hademphysema and proteinuria, one had emphysemaalone, and eight had proteinuria alone; five wereold cases of which four had emphysema andproteinuria and one had proteinuria.Of the 14 new cases, six were still employed in the

136



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/

FOLLOW-UP STUDY OF EXPOSURE TO CADMIUM FUME13

TABLE 1

NEW CASES OF CADMIUM POISONING AT FACTORY A

Urine Respiratory Function TestsCase Ae Epsr Sypos Clinical B.P. ChestNo.Agexposure ymptoms Findings (mm. Cd T.C. M.V.C. S.F.,, V C Radio- Diagnosis

Hg) Protein (gg./1.) (sec.) (1./min.) (%) (1. graphs

A 24 62 16 years near Winter cough, Normal 170/100 ± ye Cd present 0-96 145 43-6 4-75 Normal ProteinuriaCd furnace dyspnoea (not measur- (0O75)* (141) (44-3) (4-53)(1941-57) able)

A 17 64 13 years Joint pain, Rheumatoid 1130/90 + ve 30 1-16 93 35-1 3-77 Emphysema Emphysema,(1940-53) dyspnoea arthritis I(0-94) (45) (3-95) proteinuria

A 25 52 9 years Dyspnoea 2 Emphysema 130/90 + ye 30 1-21 76 I46-2 2-34 Apical Emphysema,(1946-55) years, cough (0-84) (112) (43-2) (3-7) bullae proteinuria

A 26 51 21 years near None Normal 160/70 + ye 10 0-62 173 66 3-75 Normal ProteinuriaCd. furnace (0-45) (145-5)1 (50) (4-12)(1936-57)

A 27 52 10 years None Normal I170/100 -F ye 10 0-69 125 43 4-18 Normal Proteinuria(1947-57) (0-57) (127) (44) (4-12)

A 28 48 13 years Dyspnoea on Normal 150/100 + ve 60 0-82 140 45 4-50 Normal ProteinuriaI (1944-57) Ieffort; soreness (0-54) (160) (45) (5-09)

of nose withoccasionalepistaxis

A 29 59 17 years near Dyspnoea on Obesity, 210/100 ± ye Cd present 0-74 99 41 3-45 Normal ProteinuriaCd furnace effort benign (not measur- (0-59) ( 10-4) (40) (3-95)(1936-53) hypertension able)

A 30 61 20 years near Dyspnoea on Obesity, 130/90 + ye 18 I0-77 125 45-8 3-90 Not ProteinuriaCd furnace effort, pains osteoarthritis (0-56) (3-61) radio-(1936-56) lower limbs graphted

A 31 i53 21 years Dyspnoea on Emphysema 140/90 ve i 60 1-10 143-5 51-1 4-0 Emphysema Emphysema,(1936-57) effort (0-63) 1(133-3) (49.1) (3-88) proteinuria

A 32 29 9 years near None Normal 130/90 +L ye 10 0-41 151-2 55-5 3-89 Normal ProteinuriaCd furnace (0-28) (185-6) (69-0) (3-84)(1942-53)

A 33 151 15 years near Dyspnoea on Obesity, 160/90 + ve No 0.42 109-5 1444 3-53 Normal ProteinuriaI Cd furnace effort bronchospasm cadmium (0-36) (113-6), (45-1) (3-60)

(1936-51)A 34 67 20 year-s near Cough with Emphysema, 160/90 + ve 14 1-30 44-7 17-9 3-57 Not Emphysema,

Cd furnace sputum 2 years, chronic (0-83) (123-5) (41-0) (4-30) radio- proteinuria(1936-56) dyspnoea on bronchitis graphed

effort 1 year,1pain in joints3 years

A 35 56 35 years Dyspnoea and Emphysema, 150/100 + ye 9 1-21 81-0 28-3 4-09 Normal Emphysema,(1920-55) cough bronchitis (0-68) (125-9) (39-5) (4-56) proteinuria

A 36 46 21 years Occasional Emphysema, 110/80 Normal Cd present 1-28 95-6 35-7 3-82 Normal Emphysema(1936-57) dyspnoea on bronchospasm (not measur- (0-83) (1 18-6) (45-6) (3-71)

effort able)

*Figures in brackets refer to 1953 results.

casting shop, four on the copper-cadmium furnaces,and two in the vicinity casting brass and bronze;they had done this work for between 10 and 21 years.Eight had left the casting shop at some time between1951 and 1956; their period of service in the castingshop varied between nine and 35 years. Two hadretired (Cases A30 and A34), one because ofarthritis and the other because of dyspnoea andarthritis; the remaining six worked in differentdepartments in the factory, Case A 17 because ofrheumatoid arthritis and Case A35 because of coughand dyspnoea. All 14 of these men were normal atthe time of the investigation in 1953. Four of thesemen (Cases A17, A29, A32, A33) had stoppedworking in the casting shop before that date so thatthey had not been further exposed to cadmium inthe interim period.At this factory nine cases of chronic cadmium

poisoning were known to exist amongst the 58 menoriginally in the exposed group. There were in addi-tion four men whose case records were summarizedby Bonnell in 1955 (Cases Al, A2, A3, and A4).

Of these 13 men, Case Al died from chroniic renalfailure (Bonnell, 1955), Case A4 died at home in1955 of congestive cardiac failure as the result ofemphysema, and Case A9 died at home in 1956of bronchogenic carcinoma; unfortunately nonecropsies were performed on the latter two cases.Cases A2 and A3 were seen and in both the symptomshad become so severe that disability was complete;they were unable to leave their homes unless takenby motor car. On clinical examination both hadsevere emphysema, protein was present in the urine,and the respiratory function tests were grosslyabnormal. Both men were excreting cadmium inthe urine, Case A2, 60 .tig. of cadmium per litre,Case A3, 36 ttg. of cadmium per litre. Five of theremaining eight men were re-examined and therelevant follow-up notes are included in Appendix I.The remaining three known cases of chronic

cadmium poisoning were not examined, but it wasascertained from the general practitioners concernedthat Cases A6 and Al13 were severely disabled asthe result of shortness of breath; Case A6 had

137



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/

138 BRITISH JOURNAL OF INDUSTRIAL MEDICINE

TABLE 2

NEW CASES OF CADMIUM POISONING AT FACTORY B

Case Clinical |B.P. Urine Respiratory Function Tests ChestCase Age Exposure Symptoms Findings (mm. Radio- DiagnosisNO. Findings ~~~~~~Hg) Protein Cd T.C. M.V.C. S.F70 V.C. graphs(g±g./l.) (sec.) (1./mmn.) (% (.)B 11 42 18 years Dyspnoea 3 Normal chest 150/80 + ve 28 0-73 84-8 26-4 4-58 Normal I Proteinuria

(1939-57) years, cough 3 (0 60)* (92-6) (285) (462)years

B 16 50 5 years Chronic nasal Normal chest 130/90 yve 46 0 55 181-3 59 2 4 37 Normal Proteinuria(1945-50) discharge (0-43) (195-9) (60-0) (4-66)

B 17 44 18 years None Normal chest 140/90 -+- ve No Cd 160-4 160-4 39-8 5-57 Normal Proteinuria(1939-57) I (0-63) (205 0) (53-3) (5-52)Crane driver:situated overCuCd furnaces

B 18 41 7 years None Refused - + ve No Cd 0 55 165 8 62-3 3-80 Normal Proteinuria(between clinical (0 40) (189 0) (62-7) (4 30)1935-45) examination

B 33 62 7 years Cough, sputum, Chronic 160/90 ve No Cd 0 71 61 7 35 6 2-48 Normal Proteinuria(1940-47) dyspnoea 3 bronchitis (0-48) (106-5) (52-9) (2-88)

yearsB 19 49 16 years None Normal 140/90 vye 40 0-48 150-4 57 4 3 75 Normal Proteinuria

(1941-57) (045) (1497) (532) (402)B 31 64 14 years Dyspnoea on Very small 190/100 + ve 38 0-39 92-6 81-1 1-63 Normal Proteinuria

(1936-50) effort 6 years man, other- I (0-34) (97-2) (51-2) (2 70)(intermittent) wise ncrmal

chestB 20 53 11 years Dyspnoea on Normal chest 130/90 ve 110 0-74 100-3 44-8 3-20 Normal Proteinuria

(1946-57) effort 4 years, (0-54) (119-6) (48-6) (3 50)cough 6 years

B 21 47 5 years Dyspnoea on Normal chest 130/80 + ve Cd present 1-08 96-4 32-6 4-22 Normal Proteinuria(between effort 3 years, (not measur- (0-96) (120-3), (37-1) (4 62)1940-47) cough 2 years able)

B 22 52 3 years Symptom free (Information yve Cd present - -_ _ _ Proteinuria(1939-42) gained from (not measur-

family doctor) able)

*Figures in brackets refer to 1953 results.

emphysema and congestive cardiac failure andCase A13 had emphysema. A specimen of urinefrom the former was now shown to contain protein.Case AI0 was able to continue at work.

In the previous investigation (King, 1955) atmos-pheric samples were taken from two positions in thecasting shop, one near the pit furnace where themaster alloy was manufactured, position A, and onenear the rocker furnace, position B. On this occasionall samples were taken from position B only.Samples were taken on nine consecutive periods of12 hours each in order to estimate the average con-centration during this time. On no occasion did theconcentration of cadmium in the atmosphere exceed37 tg./m3 for any one nine-hour period and theaverage concentration for the week was 13 ,ug./m3.The results compare favourably with those reportedin 1953 when the range of atmosphere concentrationwas 13 to 89 /jg./m3 in 12-hour periods. The betterconditions had been achieved by improved exhaustventilation over the rocker furnace. The traceobtained on the strip recorder suggested that partat least of the general concentration of cadmiumin the atmosphere was due to occasional bad position-ing of the exhaust hood by a particular furnace man.

Factory B.-Thirty-seven of the 42 men originallyin Group lB attended for examination. Of the fiveabsentees one man known to have chronic cadmium

posioning had died (Case B3). The other four weretraced through their general practitioners; threewere in apparent good health and one was found tohave the characteristic proteinuria. Of the 37 menseen at the factory, 18 showed signs of chroniccadmium poisoning. There are therefore 19 cases tobe discussed. Of these, 10 were new cases all withproteinuria (Table 2) and nine were old cases, fourwith emphysema and proteinuria, four with pro-teinuria alone, and one with emphysema alone.Of the 10 new cases, four continued to work on the

copper-cadmium furnaces; they had done this workfor between I1 and 18 years. The other six had notbeen exposed to cadmium in their work for at leastseven years. These 10 men were examined in 1953and found to be quite normal.

In addition, three men who had been shown tohave proteinuria on one occasion by Kench andSmith (1957) provided repeated specimens of urinebut the proteinuria could not be demonstratedunequivocally. Quantitative protein estimation bythe Biuret method was carried out on one occasionand the protein content varied between 3-5 and 5-5mg. total protein per 100 ml. of urine. It is possiblethat the proteinuria is transient when it first appears.Should that be so, careful observation and repeatedtesting of the urine would show the onset of per-sistent proteinuria in these cases.At this factory 10 cases of chronic cadmium



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/

FOLLOW-UP STUDY OF EXPOSURE TO CADMIUM FUME

poisoning were known to exist amongst the 42 menoriginally in the exposed group (Group I B, Bonnell,1955). The findings at that time were summarizedin table form and the case history of one man whowas severely disabled (Case B1) was described in full,Of these 10 men, one (Case B3) had died of acutebronchopneumonia and the post-mortem findingshave been reported by Kench and Smith (1957). Theremaining nine men were re-examined in the presentinvestigation and the relevant follow-up notes areincluded in Appendix 1.

Master Alloy Manufacture.-The master alloy(50/50 copper-cadmium) was manufactured in aseparate building from the main casting shop wherethe low percentage (1 %) copper-cadmium alloyswere cast. Atmospheric samples were taken in bothworkshops.A direct comparison with conditions existing at

the time of the previous investigation (King, 1955)was impossible owing to improvements in the exhaustsystem and pouring technique. The atmosphericsamples were taken at a position nearer to thefurnacemen than on the previous occasion. For aperiod of 10 hours of master alloy casting the averageconcentration ofcadmium was 30 ,ug./m3. During theprocess of mixing and pouring, the furnacemen woreSiebe Gorman microfilter dust respirators, theefficiency of which is discussed in Appendix 2.

Final Alloy Manufacture.-Atmospheric sampleswere taken at two positions comparable to pointsF and G in the previous investigation for eightconsecutive 12-hour periods. The results are sum-marized in Table 3.

TABLE 3RESULTS OF ATMOSPHERE SAMPLING AT FACTORY B

Position F Range 18-80 ug. Cd/m3 (29-270 !Lg. Cd/3)Average 46 ,tg Cd/m3 (106 Lg./m3)

Position G Range 10-41 ;Lg. Cd/m3 (5-56 ILg./m3)Average 20 ug. Cd/m3 (38 Lg./m3)

Figures in brackets refer to previous report (King, 1955).

These data suggest that the working conditionshave improved since the earlier survey. At that timeadditional amounts of cadmium could have beenabsorbed by the workman as the result of workingclose to the furnaces during mixing and pouringoperations. In 1957 Siebe Gorman microfilterrespirators were worn by the furnacemen duringthese operations and the atmospheric concentrationsprobably represent maximum exposure.

Cadmium Content of Respirator Pads.-Theexposure of the men to cadmium fume depended toa great extent on their care when working near the

source-of the fume. The working technique and thevariable nature of the fume emission caused seriousdifficulties when attempting to assess accurately thedegree of exposure of an individual workman. Thisproblem was reduced by the use of efficient dustrespirators, but the cadmium content of the filterpads after use gave a measure of the possibleexposure of an individual under the conditionsobserved at this factory.The filter pads from the respirators used by the

furnacemen during each casting operation were col-lected and their cadmium content estimated (Table 4).

TABLE 4CADMIUM CONTENT OF FILTER PADS OF RESPIRATORS

Master Alloy Final AlloyCasting Casting

12 filters obtained 20 filters obtainedfrom from

2 furnacemen 7 workmen

Average time of use 9 min. 20 min.Average cadmium content 0-14 mg. 0 5 mg.Range of cadmium content 0-05-0(31 mg. 0-08 -3-65 mg.

In this series three of the four highest values werefrom the pads supplied by one workman (Case B20).It is interesting to note that this man's urinaryexcretion of cadmium was far higher than that ofany of his workmates (Table 2).

These data give the total cadmium content of theair inhaled by the workmen concerned, and includethe large non-respirable particles of cadmium oxideor particles of mixed oxides from sparks. They do,however, demonstrate the possible variation inexposure of workmen performing essentially thesame work under the same conditions.

Results of Respiratory Function TestsThe number of men examined on this occasion

and the mean ages of the groups are shown inTable 5. The table also shows the mean values inthe three groups for vital capacity, maximum venti-latory capacity, swept fraction, and time constantof the expiratory fast vital capacity curve. Thecorresponding mean values obtained four years agoare also shown for comparison. In the exposedgroups, the mean values for 1957 were lower thanthe corresponding values for 1953 for vital capacity,swept fraction, and maximum ventilatory capacity,and higher for time constant. This suggested deteri-oration in respiratory function. However, the effectof age and the incidence of respiratory disease inthe population from which these samples weredrawn must be taken into account before any con-clusions are drawn as to the effects of exposure tocadmium fume.

139



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/


TABLE 5MEAN VALUES AND STANDARD DEVIATIONS FOR THE CONTROL AND EXPOSED GROUPS IN EACH FACTORY

No. of Mea Ae Vital Capacity Time Constant Swept Fraction MVC.Groups Subjects IeanAge es) (sec.) (_ (litres/mn.)

1957 1953 1957 1953 1957 1953 1957 1953 1957 1953 1957 1953

Controls:Factory A - 60 - 49 4 - 3-99 - 0-59 - 49 3 - 138

(9 30)* (0 72) (0 16) (7-75) (32)Factory B 63 44 37 7 39 8 4 13 4-46 0-64 0 58 47 3 47 5 136 148

(13 19) (12 00) (1 09) (0-69) (0-18) (0.26) (9 95) (8 35) (37) (29)

Exposed:Factory A 45 56 51 4 49-0 3 90 4 12 0-94 0 76 42 7 44-3 121 130

(9.29) (9.96) (1 06) (1 06) (0-48) (0 44) (10 84) (10-75) (47) (47)Factory B 35 40 45-4 41 4 3-91 4 32 0-97 0 79 45 9 46 2 128 142

(10 69) (10-62) (1-07) (0-82) (0-84) (0-63) (12 45) (12-29) (45) (50)

*Standard deviations are in brackets.

TABLE 6REGRESSIONS ON AGE FOR VITAL CAPACITY, TIME CONSTANT, SWEPT FRACTION, AND MAXIMUM VENTILATORYCAPACITY IN CONTROL AND EXPOSED GROUPS OF BOTH FACTORIES, WITH A COMPARISON BETWEEN SLOPES OF

REGRESSION LINES

Test Group Regression Equation S.E.b Difference in Slope between t P

V.C. Control:1957 (B) y = 592 - 0047x 0-0087 1957 and 1953 controls 1 46 0 2 > p> 0 11953 (B) y = 5 68 - 0031x 00075 J

Exposed:1957 (A) y = 7 29 - 0066x 0.014 1957 Control and exposed 1 14 0 3 > P > 0-21957 (B) y = 675 - 0063x 0 014 1957 Control and exposed 0 95 0-4 > p > 0-2

T.C. Control:1957 (B) y = 0 58 + 0-0014x 0 0017 f 1957 and 1953 controls 0-04 p > 0 91953 (B) y = 0 52 + 00015x 0-0033

Exposed:1957 (A) y = 0 02 + 0 018x 0 0074 1957 Control and exposed 2 17 0 05 > p> 0-021957 (B) y = -0-46 + 0-031 x 0012 1957 Control and exposed 2 40 0-02 > P> 0 01

S.F.,, Control:1957 (B) y = 45 0 + 0 062x 0-096 R 1957 and 1953 controls 0.91 0 4 > p > 0-31957 (B) y = 50-3 - 0069x 011 J

Exposed:1957 (A) y = 697 - 053x 0 16 1957 Control and exposed 3 14 IP < 0 011957 (B) y = 602 - 032x 0.20 1957 Control and exposed 1 72 0-1 > P > 0 05

M.V.C.70 Control:1957 (B) y = 1859 - 1 33x 0 31 f 1957 and 1953 controls 0 29 0 8 > p > 0 71953 (B) y = 195-4 - 1-20x 032 J

Exposed:1957 (A) y = 2834 - 317x 0-62 1957 Control and exposed 2 64 0-02 > p > 0011957 (B) y = 2558 - 281x 0-56 1957 Control and exposed 2 29 005 > p > 0.02

Comparison of Control Groups in 1953 and 1957.-The regression on age was calculated for each of thevariables measured in the present control group. Itwas ascertained that this relationship could beassumed to be linear over the age-range studied(Armitage, 1955). These regression equations areshown in Table 6. The regression lines obtainedwere then compared for both slope and positionwith the corresponding regression lines obtained inthe 1953 control group in Factory B.No significant differences were shown to exist

between these two control groups in the slope ofthe regression lines for time constant, M.V.C.,vital capacity, or swept fraction (Table 6). For thetime constant, the results in the two control groupswere close, there being no significant difference inslope or in position of the two regression lines (Fig.1). For the vital capacity and M.V.C., the regression

on age was similar in the two groups but the distanceapart of the regression lines was significant, the 1957values being lower than the 1953 values.

Comparison between Present Control and ExposedGroups.-This comparison was effected by the sametechnique of fitting linear regression lines to theavailable data. The regression equations obtainedfor the exposed groups in both factories are shownin Table 6.

Significant differences between the control andexposed groups were obtained for time constant,swept fraction, and maximum ventilatory capacity.It can be seen that in the exposed groups the maxi-mum ventilatory capacity decreased at a greater ratewith increase in age, and the time constant of the ex-piratory fast vital capacity curve increased at a greaterrate with age. The regression lines representing

140



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/

FOLLOW-UP STUD Y OF EXPOSURE TO CADMIUM FUMEFIG. I.-Regression of time constant on age in the control and

exposed groups of both factories.

.......FACTORY A EXPOSED_-- u 'B'. 0

-a 'B CONTROL 1957_- _ 8e of 953.

/

/ .r'

/.l r

/.//-/-.. .. . ~/,f............ function to factors other than the increase in ageand the expected incidence of respiratory disease inthe population from which this sample was drawn.The expected change in each of the tests over a

period of four years was calculated from the regres-sion equations based on the present controls. Thedifference between the 1957 and 1953 performanceof the exposed group was adjusted by this amount,so that any residual difference could be attributedto some factor other than the expected deteriorationdue to increase in age. These residual differenceswere tested to see whether they were significantlydifferent from zero, allowance having been madefor the scatter of the observations about the regres-sion line of the control subjects. For time constantthese residual differences were highly significant;they were also significantly different from zero forthe M.V.C., suggesting that the deterioration in thesequantities over four years was greater than could beexpected on the basis of age alone. For vital capacityand swept fraction, the residual differences were notsignificantly different from zero (Table 7). For thepurposes of this analysis, only those subjects workingat the time of the survey were included in the exposed

30 40 50 60AGE IN YEARS.

REGRESSION Of TIME CONSTANT ON AGE

these equations for the time constantin the control and exposed groups are

shown diagrammatically in Fig. 1.The swept fractions were also foundto decrease at a greater rate in theexposed groups with increase in age,

although the difference between theexposed group in Factory B and thecontrol group does not reach the 5%level of significance. No significantdifference could be found betweenthe control and exposed groups inthe rate at which the vital capacitydecrease with increase in age.

Comparison between 1953 and 1957Exposed Groups.-Fig. 2 shows theindividual values for time constantfor 1957 plotted against the respectivevalues for 1953. The majority ofvalues for time constant were higherin 1957, suggesting a deterioration ofrespiratory function. In the absenceof a follow-up study of the originalcontrol group it was difficult to attri-bute the deterioration in respiratory

*0 so

La

I.E

1957T.C.

1.0

I.C

0.6

ox

FIG. 2.-Comparison of individual time constants in 1953 and 1957in the exposed groups of both factories.

a

: *- :::/~~~~~~~~~~::-@.@.,/~~~~~~~~~~:: */

lr*. */:.-/~~~~

* /wA - *w s

0Q4 0QT0.8 1.0 121TlIME CONSTANT 19S3.

I.&

1.4

1.2'

T.C.

0*

a4l

141

20

t6 lie



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/

142BRITISH JOURNAL OF INDUSTRIAL MEDICINE

TABLE 7SIGNIFICANCE OF MEAN DIFFERENCES BETWEEN

1953 AND 1957 RESULTS

Test Mean ResidualDifferenceP

V.C. + 0 126 1-0 04 > p> 0 3T.C. + 0-181 5 0 0 0 < P< 0-001S.F. + 2-16 1-4 0-2 > P> 0-1M.V.C. +11-18 25 002> p> 0-01

group. In this way the control and exposed groups

were made comparable. The significant differencesshown in Table 7 were obtained despite the fact thatthe results of the men not at work, i.e., the mostseverely affected members of the exposed group,were not included.From these results, it was concluded that the

deterioration in performance over four years wassignificantly greater than could be expected fromthe behaviour of the control group studied.

Individual Performances.-In the initial investi-gation four years ago limits of normality-for-sweptfraction and time constant were defined, which werebased on the performance of the control group ineach factory. Men with a swept fraction less thantwice the standard deviation from the mean or a

time constant greater than twice the standarddeviation from the mean were selected for com-

parison with clinical and radiological findings. Inthe present investigation the same values for thelimits of normality were adopted in order that resultsmight be compared.Of the 21 men who gave abnormal results in 1953,

two have died (A4, B3) and have been consideredelsewhere in this paper. Four men from Factory Ahad left the company's employ and were not re-examined (A10, A13, A14, A18). It was ascertainedthat two of these men were suffering from shortnessof breath. The remaining 15 men again had valuesfor swept fraction or time constant outside thedescribed limits. These values are shown in Table 8.

In the 1957 survey an additional 11 men hadabnormal respiratory function tests; in four menboth the swept fraction and the time constant wereabnormal, in six men only the time constant and inone man only the swept fraction were abnormal(Table 8). Of the 10 men whose time constants weregreater than two standard deviations from the meanfor the first time in 1957, there were seven whosetime constants fell between one and two standarddeviations from the mean in 1953.

Results of Examination of UrineNine of the 46 men in Factory A were found to

give a consistently positive test for glucose in theurine. Ten of the 37 men in Factory B gave positive

TABLE 8SWEPT FRACTIONS AND TIME CONSTANTS FALLINGOUTSIDE TWICE THE STANDARD DEVIATION OF THEMEAN FOR CONTROL GROUP IN FACTORIES A AND B

Case No. Age T.C. S.F.,,>0-91sec.* <3YA2 56 2 59 22-4A3 62 3 05 22-2A5 62 1-28 Did not performA8 59 1-44 33-0All 69 2-12 28-7A12 58 2 95 15-2A15 49 1 69 29-3A16 54 1-31 -A17 64 1-16 -A20 44 1126 -A24 62 0-96 -A25 52 1-21 -A31 53 1-10 -A34 67 1-30 17 9A35 56 1-21 28-3A36 46 1-28 -A37 34 1139 31 7A38 47 1-25 -A39 56 0-93 -

A40 60 - 27-3

Bl 56 4-55 Did not performB2 53 1-64 -

B4 61 2-10 27-4B5 63 2-17 304B6 54 3-15 21-2B1l 42 - 26-4

Subjects Al I and A24 to A40 gave values within the normal limitsin the initial survey.

*2 S.D.s from control mean.

results with the same test. This slight degree ofglycosuria was consistently present in repeatedspecimens of urine. Two men at Factory B (CasesB2 and B4) were found to have 0-25% and 1%respectively of glucose in the urine. Both these menhad typical mild diabetic glucose tolerance curves.Seven of the nine men exhibiting glycosuria inFactory A also had proteinuria and all 10 of themen at Factory B with glycosuria also had pro-teinuria.The results of analysis of cadmium excretion in the

urine are given in Tables 1 and 2. No correlationwas found between the excretion of cadmium in24-hour specimens and the excretion in spot speci-mens of urine.

Urinary cadmium excretion in the control groupwas not studied on this occasion. Bonnell (1955)gave data of a similar control group of 89 men,76 of whom were not excreting detectable amountsofcadmium, 10 were excreting 10/30 ,ug. ofcadmiumper day, and three were excreting over 30 ,ug. perday. On that occasion the polarographic techniqueused had a similar sensitivity, i.e., 10 jig. cadmiumper litre of urine, to that used in the survey nowreported. Smith and Kench (1957) found a similardistribution amongst the 28 subjects in their controlgroup.As in 1953 questioning did not reveal evidence of

watery discharge from the nose or loss of the senseof smell. It was impossible to assess whether a

142



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/


yellow discoloration of the teeth was presentowing to the poor oral hygiene in the majority ofthe workers. The blood pressure readings were nothigher than in the control groups; in fact they werealmost identical in individual workmen after afour-year interval. No cadmium was detected inthe blood obtained by venepuncture.

DiscussionIt is clear from the evidence obtained as the result

of this investigation that the first sign of chroniccadmium poisoning can occur after a long latentinterval following cessation of exposure to cadmium.When signs or symptoms of poisoning are present,the disease progresses in severity with time and thisdeterioration occurs in the absence of furtherexposure to cadmium.

It can be seen from Table 2 that six of the 10 newcases at Factory B had ceased to be exposed tocadmium by 1950. One of these men had only beenexposed for three years, from 1939 to 1942. All werenormal on examination during the survey carriedout at this factory in 1953 (Bonnell, 1955). AtFactory A four of the 14 new cases had their lastexposure in 1953. Since working conditions weregood in 1957, as evidenced by the environmentalstudy, it is unlikely that any of the 24 new cases weredue to absorption of cadmium in the period 1953 to1957.The men diagnosed as having chronic cadmium

poisoning in 1953 showed evidence of deterioration.Three had died (Cases A4, A9, and B3), two as theresult of respiratory disease and one of broncho-genic carcinoma. At Factory A only two of theknown cases of chronic cadmium poisoning werestill at work (Cases A7 and A8), and in one of these(Case A8) there was undoubted evidence of increasedseverity of the disease. One man (Case A6) who hademphysema only in 1953 had become completelydisabled and for the first time had proteinuria. AtFactory B only four of the 10 men known to havechronic cadmium poisoning continued at their usualwork in the casting shop. Deterioration was demon-strated by clinical examination and respiratoryfunction tests in the remaining six men. One wascompletely disabled, another was only capable ofcarrying out sedentary work, and four were onlyable to do light work.The results of the respiratory function tests showed

a deterioration with increase in age in the exposedgroups compared with the control group.A new group of involuntary controls was investi-

gated at one of the factories; the results are com-parable with those obtained in volunteer controlgroups in 1953, thus validating the original controlseries and verifying that the procedure adopted in

1957 did not vary significantly from that adopted in1953.The measurement of the time constant of the

expiratory fast vital capacity curve gave unequivocalresults. No difference could be demonstrated betweenthe performance of the 1953 and 1957 control groups,as shown by the similarity of the regression equationson age in each group. However, there was a signi-ficant difference between the 1957 control group andthe exposed groups in each factory. This differenceis well illustrated in Fig. 1. The results for theM.V.C.70 and for the swept fraction support thesefindings. No useful conclusion could be drawn fromthe vital capacity results in this any more than in theinitial survey. The expiratory fast vital capacitycurve provided a reliable and sensitive index ofventilatory function.The comparison between the exposed groups in

1953 and 1957 showed a greater deterioration thancould be expected from the behaviour of the controlgroups studied. The deterioration over four yearsin the control group was calculated from obser-vations made on a single occasion. The 1957investigation was handicapped in this respect becausethe original control group was not available forexamination.

In assessing the respiratory function of individuals,the results of these tests corresponded well withclinical and radiological findings. In a number ofcases men who gave abnormal results on this surveygave readings which were close to the normal limitsfour years previously. It would be of interest torepeat these tests after a further period on menwhose results were for the first time close to thenormal limits.The radiological findings agreed closely with the

clinical findings and the results of the respiratoryfunction tests. The chest radiographs were inspectedwithout any knowledge of the clinical findings sothat an unbiased opinion could be taken of eachchest film.

All 10 new cases at Factory B and eight of the 14new cases at Factory A had proteinuria only. Ithas been established that the proteinuria in the newcases is identical with that described by Kekwick(1955). Spectrophotometric examination of theurinary proteins gave extinction values at 280 muL,in the range found for proteins, providing additionalevidence that they are protein in nature. Estimationof the bound carbohydrate of these proteins by themethod of S0rensen and Haugaard (1933) on threesamples of the proteins gave the following values,18 2 %, 22-3%, 29-5 %. These figures indicate thatthese are mucoproteins with a high polysaccharidecontent and this is consistent with their reaction tosulphosalicylic acid (Kekwick, 1957).

143



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/


Proteinuria is a common finding in renal disease;in chronic glomerulonephritis, pyelonephritis, orthe kidney of hypertension the proteins are plasmaproteins with an excess of albumin. The protein inthe urine of men suffering from cadmium poisoningis a specific protein which has a molecular weightof 20,000 to 30,000. It is not clear therefore whetherthis protein is present in the urine as the result ofrenal damage or whether it has a sufficiently smallmolecule to pass through a normal kidney. Since,however, this proteinuria only occurs followingabsorption of cadmium it is evident that a toxicprocess has occurred. The only other naturallyoccurring disease in which a specific proteinuriaoccurs is multiple myeloma, when Bence-Jones pro-tein is present in the urine of a certain proportion ofcases. Patients suffering from multiple myeloma candie as the result of renal failure, the histological lesionbeing essentially one of primary tubular damage.

Clarkson and Kench (1956) studied the amino-acid content of the urine of men exposed to lead,mercury, uranium, and cadmium, the latter ascadmium oxide dust in the accumulator industry.They found that the cadmium and uranium workershad an increased amount of amino-acids in the urineand that this finding was more marked in thecadmium workers than in those exposed to uranium.It was particularly marked in those men who wereexcreting the specific cadmium protein. Amino-aciduria occurring in the absence ofadvanced hepaticfailure is a manifestation of renal tubular damage,provided the plasma amino-acid levels are notincreased. The plasma amino-acid levels in workersexposed to cadmium showing the aminoaciduriaare not available.During the survey described in this paper it was

found that, using a sensitive qualitative test forglucose 19 of the 82 men had glycosuria. One ofthese had suffered from diabetes mellitus for manyyears; of the remaining 18 men, 17 were alsoexcreting the cadmium protein in the urine. Thisobservation has been confirmed independently byKench and Smith (1957) during routine testing ofthe urine of workers exposed to cadmium oxide dustin the alkaline accumulator industry. These two find-ings are compatible with early renal tubular damage.Of the cases studied in hospital, one man died in

uraemia from chronic renal failure (Case Al,Bonnell, 1955) and granular contracted kidneyswere found at necropsy. The histological changeswere non-specific but in cases of advanced renaldisease it is frequently impossible to find histologicalevidence incriminating any one aetiological factor.Abnormal renal function tests have been demon-strated in three other cases (A2, B1, and B6, Bonnell,1955).

Severe renal damage can be produced in ratsfollowing repeated administration of small quantitiesof cadmium for periods of up to one year (Bonnell,King, and Ross, in preparation). The lesion isprimarily one of tubular damage, the glomeruliremaining intact. In these experiments it was shownthat if the administration of cadmium was stoppedafter four to five months, when only a slight degreeof renal damage had been produced, the kidneys ofthese animals were as severely affected at the con-clusion of the experiment as those of the animalswhich had received the cadmium for 12 months.Tubular necrosis has also been produced in rabbitsby Dalhamn and Friberg (1957) after daily sub-cutaneous injections of cadmium for 10 weeks.Knowledge will not be complete until furtherhuman cases have been studied at necropsy, but theevidence available at present suggests that progres-sive renal damage does occur following prolongedabsorption of cadmium.There is some evidence that the urinary excretion

of cadmium may be affected by the general health ofthe patient. In one instance (Case B6) the cadmiumcontent in the urine rose from 24 ,tg. per day to440 ug. per day coinciding with an acute illness, inthis case a spontaneous pneumothorax. In anothercase (A2) cadmium excretion reached a figure of900 ,tg. per day during acute bronchopneumoniabut when last examined this patient was excreting25 ,tg. cadmium per litre of urine. Case B3 whenseen in 1953 was excreting 1-24 mg. of cadmium perday in the urine. At post-mortem examination onthis patient (Smith, Kench, and Smith, 1957) the bilecontained 24 ,tg. of cadmium per g. representing anexcretion of about 19 mg. of cadmium per day. Theliver contained 160 pg. of cadmium per g. and at thisrate of excretion all the cadmium would have beenremoved from the liver in three to four months.High concentrations of cadmium were also presentin the aorta and main vessels in this case. This mandied of acute bronchopneumonia and it could wellbe that cadmium was released from the tissues duringthis acute illness.The analysis of the cadmium concentration of

the atmosphere in both factories suggested that pro-vided individual workmen took great care, therewas probably little hazard at that time. Despite this,the measurement of cadmium on face masks wornfor a single operation by men employed at FactoryB showed that even under those conditions largeamounts of cadmium could have been absorbed bya careless workman and suggested that the ventila-tion system could be improved further.Of the 100 men first examined in 1953 at these two

factories, 43 are now known to have symptoms orsigns of the disease. In addition, seven other cases

144



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/


are known among ex-employees. The disease ischronic and in the majority of cases patients maysuccumb to anoxia due to emphysema or to someintercurrent pulmonary infection before the renallesion has had time to develop. The new casesdescribed in this paper have in many instances noevidence of pulmonary involvement, so that in themthe renal lesion may become the more importantfeature of the disease.

We are indebted to the medical and personnel depart-ments of the two factories concerned for their continuedassistance and wholehearted cooperation; to Dr. DonaldHunter and Dr. P. L. Bidstrup for their advice andencouragement; to Dr. R. A. Kekwick for advice andassistance in protein identification; to Dr. Richard Dollfor his advice on the planning of the survey; and toMiss J. Peal and Mr. B. Biles for technical assistance.We should also like to thank Dr. E. H. Evison for

consultations over Cases Dl and D2, and Dr. G. L.Manson who arranged for glucose tolerance tests inCases B3 and B4 and renal function tests in Case B6.

REFERENCESArmitage, P. (1955). Personal communication.Bernstein, L., D'Silva, J. L., and Mendel, D. (1952). Thorax, 7, 255.Bonnell. J. A. (1955). Brit. J. industr. Med., 12, 181.

, King, E., and Ross, J. H. In preparation.Clarkson, T. W., and Kench, J. E. (1956). Biochem. J., 62, 361.Dalhamn, T., and Friberg, L. (1957). Acta. path microbiol. scand.,

40, 475.D'Silva, J. L., Freeland, D. E., and Kazantzis, G. (1953). Thorax, 8,

303., and Kazantzis, G. (1954). Ibid., 9, 128.

Friberg, L. (1950). Acta med. scand., Suppi. 240.Gandevia, B., and Hugh-Jones, P. (1957). Thorax, 12, 290.Kazantzis, G. (1956). Brit. J. industr. Med., 13, 30.Kekwick R. A. (1955). Ibid., 12, 196.

(1957). Personal communication.Kench, J. E. and Smith, J. C. (1957). Personal communication.King, E. (1955). Brit. J. industr. Med., 12, 198.Smith, J. C., and Kench, J. E. (1957). Ibid., 14, 240.____ ____ and Smith, J. P. (1957). Ibid., 14, 246.Sorensen, M., and Haugaard, G. (1933). C.R. Lab. Carlsberg 19,

No. 12.Wright. B. M. (1954). Brit. J. industr. Med., 11, 284.

APPENDIX 1Follow-up Notes of Cases Previously Reported

Factory A.-The following cases were investigatedfrom Factory A.

Case AS.-A man, aged 62, had cast copper-cadmiumalloys from 1938 to 1950. He continued to work in thecasting shop until March, 1954, when he was forced toretire because of the severity of his symptoms. He wasextremely short of breath, but complained of no coughand little sputum; on clinical examination there wasevidence of severe emphysema with bronchospasm. Theblood pressure was 160/110 mm. Hg and the urine con-tained protein. The respiratory function tests weregrossly abnormal (Table 8).

Case A 7.-A man, aged 60, had continued to work inthe casting shop but had not actually cast copper-cadmium alloys since August, 1956. He was symptom-free; on clinical examination blood pressure was 120/80mm. Hg, no emphysema was demonstrable, but theliver was enlarged and palpable three fingerbreadthsbelow the right costal margin. The urine contained thecharacteristic protein and traces of cadmium (10 to 15,sg. per litre). Glycosuria was demonstrable. Respiratoryfunction tests showed no deterioration since 1953 andthe chest radiograph was normal.

Case A8.-A man, aged 59, had cast copper-cadmiumalloys from 1921 to 1943 and since that time he hadworked on the manufacture of copper wire. Hissymptoms were more severe and he was constantly shortof breath aggravated each winter by cough and sputum.On clinical examination there was evidence of severeemphysema, blood pressure was 130/90 mm. Hg, andthe urine contained the characteristic protein. A mid-daysample of urine contained 70 jsg. of cadmium per litreand the test for glucose was positive. There was radio-logical evidence of increased severity of the emphysemacompared with 1953; the range of diaphragmatic move-

ment was reduced and there was an increase of thereticular pattern of the lungs with bullous areas at bothapices and at the right base. The respiratory functiontests were grossly abnormal (Table 8).

Case All.-A man. aged 69, had worked for 20 yearsin the vicinity of the copper-cadmium furnaces. In 1953there was clinical evidence of chronic bronchitis andemphysema but respiratory function tests and a chestradiograph were normal. Proteinuria was demonstrable.In March, 1955, he had a cardiac infarction and wasforced to retire from work. He was very short of breathand also suffered from a cough, sputum, and angina ofeffort. On clinical examination there was evidence ofemphysema and chronic bronchitis but in addition therewere signs of left ventricular failure with cardiac enlarge-ment, ventricular extrasystoles, and a blood pressure of160/100 mm. Hg with pulsus alternans. The character-istic protein was present in the urine and respiratoryfunction tests were grossly abnormal (Table 8) con-firming a diagnosis of emphysema. The urine contained10 /sg. of cadmium per litre.Case A12.-A man, aged 58, had worked in the

vicinity of the copper-cadmium furnaces from 1935 to1945. He had retired from work in 1953 because of theseverity of his symptoms. He complained of severeshortness of breath with "tightness" of the chest but nocough or sputum. There was clinical evidence of severeemphysema with bronchospasm, and the urine containedthe characteristic protein but no cadmium was detected.The respiratory function tests (Table 8) confirmeddeterioration.

Factory B.-The following cases were seen at Factory B.Case BJ.-A man, aged 56, had cast copper-cadmium

alloy from 1930 to 1950. In December 1956, he com-plained of a sudden severe pain over the left lower part

145



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/


of the chest which was accompanied by extreme dyspnoea.He had been extremely dyspnoeic since that time. Onclinical examination there was evidence of advancedemphysema with bullae at both lung bases. The chestradiograph confirmed this. It is probable that he suffereda spontaneous pneumothorax in December, 1956. Therespiratory function tests were grossly abnormal (Table8). Protein was present in the urine, which also contained25 tg. of cadmium per litre.

Case B2.-A man, aged 53, had continued to supervisethe casting shop. He complained of dyspnoea on effortand winter cough; emphysema was demonstrableclinically and radiologically and the respiratory functiontests were abnormal (Table 8). Proteinuria and glycosuriawere demonstrated. Glucose tolerance tests showed thetypical response of mild diabetes mellitus. He wasexcreting 20 ,ug. of cadmium per litre of urine.

Case B4.-A man, aged 61, had not been exposed tocadmium after 1953. Dyspnoea was more severe than atthe time of the previous examination; a marked degreeof emphysema accompanied by bronchospasm wasevident on clinical examination. The respiratoryfunction tests confirmed the clinical impression ofdeterioration (Table 8). Proteinuria and glycosuria weredemonstrated. On further investigation he was shown tohave a diabetic glucose tolerance curve. No cadmiumwas detected in the urine.

Case BS.-A man, aged 63, had not been exposed tocadmium after 1947. Shortness of breath was as severeas at the first examination, but he had not complained ofcough. The chest radiograph and respiratory functiontests confirmed the diagnosis of emphysema and therespiratory function tests (Table 8) showed that theemphysema was more severe. Proteinuria was presentand also traces of cadmium in the urine.

Case B6.-A man, aged 54, was exposed to cadmiumfor seven years from 1945 to 1952. On examination in1953 he was found to have emphysema and broncho-spasm; the liver and spleen were palpable on examinationof the abdomen. He remained reasonably well untilAugust, 1956, when he was admitted to hospital becauseof a sudden exacerbation of dyspnoea accompanied bysevere chest pain. On admission he was found to havea left-sided pneumothorax with collapse of approxi-mately 30% of the lung. He made an uneventful recoverywith symptomatic treatment. No protein was demonstr-able in the urine but excretion and concentration testsof the urine showed slight impairment of renal function.On examination respiratory function tests showed markeddeterioration in four years (Table 8). The chest radio-graph showed an increase in size of the hilar vessels andof the transverse diameter of the heart compared withthe film taken in 1953.The daily urinary excretion of cadmium in this man

was 24 ,ug. in 1953. During his illness in August andSeptember, 1956, this figure rose to 440 jug. Duringconvalescence at home, but before returning to work,the daily excretion was 153 pg. In June, 1957, by whichtime he had completely recovered from the acute effectsof the pneumothorax and was back at work, the daily ex-cretion was again low, namely 34 ,tg. ofcadmium per day.

Case B7.-A man, aged 41, was symptom-free, but pro-teinuria and glycosuria were present on testing repeatedsamples of urine. The urine contained no cadmium.

Case B8.-A man, aged 40, complained of somedyspnoea on exertion but no abnormality was detectedon examination. Proteinuria was again marked andglycosuria was also demonstrable. No cadmium wasdetected in the urine.

Case B9.-A man, aged 65, no longer worked in thecasting shop and was semi-retired. He was symptom-free but proteinuria and glycosuria were present inrepeated specimens of urine, which also contained 25 ,tg.of cadmium per litre.

Case BJO.-A man, aged 45, complained of somedyspnoea on effort, but no emphysema was demonstrableon examination. Proteinuria was demonstrated but noglycosuria. The urinecontained 24jg. ofcadmium per litre.

New Cases.-These cases have been seen since thereport of the previous investigation.

Case DJ.-A man, aged 51 years, in whom a diagnosisof emphysema with bronchospasm was made. Thecharacteristic protein was present in the urine. He hadworked from 1933 to 1939 casting copper-cadmiumalloys at Factory A. He served in the armed forces from1939 to 1942 when he was discharged to carry outessential industrial work; he continued casting copper-cadmium alloys from 1942 to 1945. Since 1945 he hadworked as a bus conductor.He was quite well until 1945 when he first became

short of breath on exertion with "tightness" of the chestand a slight unproductive cough. The dyspnoea graduallyincreased in severity and had become severe since 1951following an acute attack of bronchitis. Dyspnoea hadbeen constant since 1954.On examination in May, 1957, there was a slight degree

of cyanosis but no finger clubbing. The chest was fixedin inspiration. There was complete absence of cardiacand liver dullness on percussion and the heart soundswere best heard over the xiphistemum. The chest washyperresonant on percussion, breath sounds were vesi-cular, and many high-pitched rhonchi were heard. Theurine contained the characteristic protein and 12 jig. ofcadmium per litre.

Case D2.-This was a man aged 63 years who hadproteinuria only. He had worked from 1939 to 1956 inthe casting shop at Factory A casting brass, bronze, andcopper-cadmium alloys. He had refused to take part inthe survey in 1953, but did provide an early morningspecimen of urine which at that time was normal.For 12 months he had complained of dyspnoea on

exertion with cough and sputum for nine to 10 months.On examination in May, 1957, he was not cyanosed andthe fingers were not clubbed. Chest expansion was 1Iinches. Cardiac and liver dullness were within normallimits on percussion of the chest. The breath sounds werevesicular, air entry was good at both the apices and basesof the lungs, and occasional rhonchi, dispersed by cough-ing, were heard. There was no evidence of emphysema.The characteristic protein was present in the urine,

which contained 10 ,ug. of cadmium per litre.

146



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/


APPENDIX 2Dust respirators should not be used as the primary

protection against dusts or fume, but only as additionalsafeguards when the nature of the work does not allowadequate exhaust ventilation covering the source ofemission at all times. During the manufacture of copper-cadmium alloys the workmen are required to be close tothe source of cadmium oxide fume and respirator pro-

tection at these times would reduce their exposure to thefume.The effective use of respirators requires that they

should be efficient against the type of material concerned;that they should be sufficiently comfortable under con-

ditions of temperature and exertion so as not to reducethe efficiency of the workmen; and that supervision isavailable to ensure their use at the appropriate times.These conditions appeared to be satisfied in Factory Bby the Siebe Gorman "microfilter" dust respirators. Inorder to obtain data on their efficiency in the workingconditions observed against cadmium oxide fume tworespirators were obtained after their use by two workmenduring a series of mixing and casting operations.

100.

501

401

301

201

I0

0

FILTER

*1

I

The wadding filters of the respirators were each dividedinto six approximately equal sections from front to rearof the mask, and analysis for cadmium content wasperformed on each section. In order to avoid errorsarising from sparks trapped in the first sections, the dataplotted in Fig. 3 give the amount in any one section ofeach mask as a percentage of the total amount in thatsection and those to the rear of it. Thus the higherapparent efficiencies of the first two sections were pos-sibly caused by the presence of large particles, but thenext three each showed an efficiency of about 50 % whichmay be assumed to be against the residual fume after theremoval of particles formed by sparks.The data obtained thus showed that each section of

the filters contained about 50% of the cadmium oxidefume entering it. The overall efficiency is therefore thatof six sections in series, each of 50% efficiency, to give avery high overall efficiency. These respirators can there-fore be accepted as suitable protective devices to be usedin addition to exhaust ventilation covering the sources ofthe cadmium oxide fume.

a x 100c+bc+d+etf

-2= b xlOO x3= ' Xoo XO

oxidefmo nmirfle

dust respirators.

147



.bmj.com

/B


.16.2.135 on 1 April 1959. D

ownloaded from

http://oem.bmj.com/

Documents

STUDY EXPOSED TO CADMIUM OXIDE FUME - oem.bmj.com · of each specimen of urine received was estimated by polarography. The demonstration of protein in the urine is an objective test